Study On Optimal Wavelet Basis For Gas Pipeline Leakage Detection

Gas pipeline has been widely used in national economic movement as a major mode of energy transportation.Pipeline leakage will directly cause hidden dangers.Timely detection of leakage is key to ensure pipeline transportation.Supported by Natural Science Foundation of Inner Mongolia Autonomous Region-Construction of acoustic emission theoretical model for Gas pipeline leakage and study on optimal wavelet basis construction(2021LHMS05027),this study studied the wavelet basis used in gas pipeline leakage acoustic emission fault detection.The study aims at selecting and constructing suitable wavelet basis function for gas pipeline leakage detection.To this end,the following research work has been carried out.1)Selection of optimal general wavelet basis for acoustic emission signal of gas pipeline leakage.Combined with the noise reduction ability,similarity and energy-entropy criterion of wavelet,the leakage AE signal is denoised by using the improved threshold function.According to the noise reduction ability of the selected general wavelet basis,the similarity between the wavelet basis and the leakage AE signal is compared,and the approximate component energy of the leakage AE signal and the energy spectrum distribution of the signal are analyzed.The optimal general wavelet basis is selected for the leakage AE signal.2)The wavelet basis matching acoustic emission signal of gas pipeline leakage is constructed.Firstly,the scale coefficients of wavelet are parameterized by using the scale function equations constrained by the length of the wavelet filter sequence,and the scale coefficients of the length N=4 and N=6 of the wavelet filter sequence are solved,which satisfies the orthogonal wavelet constraint conditions.Secondly,according to the construction wavelet method matching the given signal,the analysis of the construction wavelet projected most of the energy of the leakage AE signal into the wavelet scale subspace,and verified the matching between the construction wavelet and the leakage AE signal.3)Compare and analyze the detection ability of the selected wavelet basis and the constructed wavelet basis on pipeline leakage.Based on the time-frequency domain analysis method of wavelet,the time-frequency domain feature recognition of leakage AE signals is studied,and the leakage frequency components of pipeline leakage faults are extracted.Choi-Willarms time-frequency domain joint distribution was used to calculate the proportion of energy of leakage fault signal,and the proportion of detection of leakage fault frequency components by different wavelet basis was analyzed.The results show that compared with the selected general wavelet basis,the constructed wavelet can not only detect the leakage fault clearly in the time-frequency domain,but also increase the proportion of the leakage fault frequency component.At the same time,the constructed wavelet basis function has a higher fault energy specific ratio for the acoustic emission signals of different pressure leakage generated by different leakage holes of the pipeline,which verifies the effectiveness of the constructed wavelet basis for detecting the leakage fault of the gas pipeline.This research has important theoretical significance and practical value to promote the development of wavelet detection of pipeline leakage AE fault signal.